On Tue, Oct 19, 2021 at 4:43 PM Marc Zyngier <maz@xxxxxxxxxx> wrote:
>
> On Sat, 09 Oct 2021 03:20:18 +0100,
> Florian Fainelli <f.fainelli@xxxxxxxxx> wrote:
> >
> > Provide the platform device mapping to the interrupt controller node to
> > the of_irq_init_cb_t callback such that drivers can make use of it.
> >
> > Reviewed-by: Rob Herring <robh@xxxxxxxxxx>
> > Signed-off-by: Florian Fainelli <f.fainelli@xxxxxxxxx>
> > ---
> > drivers/irqchip/irqchip.c | 2 +-
> > drivers/irqchip/qcom-pdc.c | 3 ++-
> > drivers/of/irq.c | 2 +-
> > include/linux/of_irq.h | 5 ++++-
> > 4 files changed, 8 insertions(+), 4 deletions(-)
> >
> > diff --git a/drivers/irqchip/irqchip.c b/drivers/irqchip/irqchip.c
> > index 3570f0a588c4..289784eefd00 100644
> > --- a/drivers/irqchip/irqchip.c
> > +++ b/drivers/irqchip/irqchip.c
> > @@ -55,6 +55,6 @@ int platform_irqchip_probe(struct platform_device *pdev)
> > if (par_np && !irq_find_matching_host(par_np, DOMAIN_BUS_ANY))
> > return -EPROBE_DEFER;
> >
> > - return irq_init_cb(np, par_np);
> > + return irq_init_cb(np, par_np, pdev);
> > }
> > EXPORT_SYMBOL_GPL(platform_irqchip_probe);
> > diff --git a/drivers/irqchip/qcom-pdc.c b/drivers/irqchip/qcom-pdc.c
> > index 173e6520e06e..819a93360b96 100644
> > --- a/drivers/irqchip/qcom-pdc.c
> > +++ b/drivers/irqchip/qcom-pdc.c
> > @@ -359,7 +359,8 @@ static int pdc_setup_pin_mapping(struct device_node *np)
> > return 0;
> > }
> >
> > -static int qcom_pdc_init(struct device_node *node, struct device_node *parent)
> > +static int qcom_pdc_init(struct device_node *node, struct device_node *parent,
> > + struct platform_device *pdev)
> > {
> > struct irq_domain *parent_domain, *pdc_domain, *pdc_gpio_domain;
> > int ret;
> > diff --git a/drivers/of/irq.c b/drivers/of/irq.c
> > index 352e14b007e7..18f3f5c00c87 100644
> > --- a/drivers/of/irq.c
> > +++ b/drivers/of/irq.c
> > @@ -538,7 +538,7 @@ void __init of_irq_init(const struct of_device_id *matches)
> > desc->dev,
> > desc->dev, desc->interrupt_parent);
> > ret = desc->irq_init_cb(desc->dev,
> > - desc->interrupt_parent);
> > + desc->interrupt_parent, NULL);
> > if (ret) {
> > of_node_clear_flag(desc->dev, OF_POPULATED);
> > kfree(desc);
> > diff --git a/include/linux/of_irq.h b/include/linux/of_irq.h
> > index aaf219bd0354..89acc8b089f0 100644
> > --- a/include/linux/of_irq.h
> > +++ b/include/linux/of_irq.h
> > @@ -9,7 +9,10 @@
> > #include <linux/ioport.h>
> > #include <linux/of.h>
> >
> > -typedef int (*of_irq_init_cb_t)(struct device_node *, struct device_node *);
> > +struct platform_device;
> > +
> > +typedef int (*of_irq_init_cb_t)(struct device_node *, struct device_node *,
> > + struct platform_device *);
>
> Having added some type-checking to the IRQCHIP_MATCH() #definery, I
> end-up with warnings such as:
>
> In file included from ./include/linux/kernel.h:16,
> from drivers/irqchip/irq-bcm7038-l1.c:12:
> ./include/linux/minmax.h:20:28: warning: comparison of distinct pointer types lacks a cast
> 20 | (!!(sizeof((typeof(x) *)1 == (typeof(y) *)1)))
> | ^~
> ./include/linux/irqchip.h:41:2: note: in expansion of macro ‘__typecheck’
> 41 | __typecheck(typecheck_irq_init_cb, fn) ? fn : fn
> | ^~~~~~~~~~~
> ./include/linux/irqchip.h:44:12: note: in expansion of macro ‘typecheck_irq_init_cb’
> 44 | .data = typecheck_irq_init_cb(fn), },
> | ^~~~~~~~~~~~~~~~~~~~~
> drivers/irqchip/irq-bcm7038-l1.c:459:1: note: in expansion of macro ‘IRQCHIP_MATCH’
> 459 | IRQCHIP_MATCH("brcm,bcm7038-l1-intc", bcm7038_l1_of_init)
> | ^~~~~~~~~~~~~
>
> Clearly, you didn't update the drivers you just converted to
> IRQCHIP_MATCH(), and only fixed the QC driver.
>
> FWIW, I'm planning to take something like the hack below into the tree
> to detect this sort of stuff early.
>
> M.
>
> diff --git a/include/linux/irqchip.h b/include/linux/irqchip.h
> index ccf32758ea85..146a9d80a6a2 100644
> --- a/include/linux/irqchip.h
> +++ b/include/linux/irqchip.h
> @@ -33,7 +33,15 @@ extern int platform_irqchip_probe(struct platform_device *pdev);
> #define IRQCHIP_PLATFORM_DRIVER_BEGIN(drv_name) \
> static const struct of_device_id drv_name##_irqchip_match_table[] = {
>
> -#define IRQCHIP_MATCH(compat, fn) { .compatible = compat, .data = fn },
> +/* Undefined on purpose */
> +int typecheck_irq_init_cb(struct device_node *, struct device_node *,
> + struct platform_device *);
> +
> +#define typecheck_irq_init_cb(fn) \
> + __typecheck(typecheck_irq_init_cb, fn) ? fn : fn
That's nice! Shouldn't it also be used for IRQCHIP_DECLARE?
> +#define IRQCHIP_MATCH(compat, fn) { .compatible = compat, \
> + .data = typecheck_irq_init_cb(fn), },
I was going to say I wasn't really a fan of IRQCHIP_MATCH given it had
nothing irqchip specific about it, but you fixed that now...
>
> #define IRQCHIP_PLATFORM_DRIVER_END(drv_name) \
> {}, \
>
> --
> Without deviation from the norm, progress is not possible.
